And the truth (or at least, an attempt at redressing the balance)

Talking on Topic: I try Menshn so you don’t have to

Menshn is a bold idea. Aiming to replace Twitter is no small thing (I say “replace” because the site doesn’t augment existing use of Twitter but offers a new service entirely) and it’s a brave move to do. It perhaps goes without saying that the project would’ve already died on its arse without the high-profile involvement of UK MP Louise Mensch (who isn’t the namesake, apparently…). For that reason it makes sense that the site initially launched in the US — early adopters would be unlikely to have strong opinions on Mensch herself and initial bugs could be ironed out before the more scathing British public were let loose on it.

And what initial bugs they were. Twitter was ablaze a few weeks ago with criticisms of both the site and its technical creator, Luke Bozier. Turned out the site had a bunch of (hotly denied) security holes, mostly quite basic errors. After these had been fixed and the site moved to a new host, I decided to take a look myself.

Crimes against typography

The first thing I was struck by was the quite ugly design. This example is perhaps unfair since it’s presumably not how the content was supposed to look, but the leading on the quote text is awful:

User interface? We don’t need no stinkin’ user interface!

Ignoring this hideousness, I registered for the site. The confirmation email immediately fell into my “suspected spam” box in Gmail, which was mildly interesting. I activated my account and logged in to be confronted with this toolbar at the top of every page:

This impenetrable interface  gives no clues what the three mysterious icons next to ‘home’ indicate, and hovering over them offers no clues either. Clicking around I managed to figure out what they refer to but it seems like a basic UI fail to not make clear what textless icons actually refer to.

Chronological context is for suckers

I thought I’d move on to some actual “menshns” (I’m still not sure if this is their equivalent to “tweets”). Here’s a random conversation from the front page:

It took me more experimental clicking and hovering to establish that none of these conversations have any indication of time/date. Perhaps this is intentional but I found it irritating and confusing to work out the sequence of conversation, or indeed the time gaps between each one. How can I judge whether a conversation is worth participating in if I have no idea how long ago the last post was?

Talk on topic… if you’re a search engine

Giving up on this, I decided to enter one of the “chatrooms”. I clicked “judaism”, mostly by accident. Here’s what I got:

While I could clearly see a conversation was taking place (again, no timestamps) though, it was a little tricky to figure out what the topic was beyond the single-word title. For this example perhaps “judaism” is enough to go on, but one of the chatrooms linked from the homepage is simply called “why”. I was confused — when creating a room you’re asked to enter a description. Examining the page’s source code revealed this:

Right there in the (invisible to humans) meta tags, we have the room’s description. Why isn’t this anywhere on the page? Surely if the aim is to “talk on topic”, it might help to, you know, tell people what the topic is.

Accessibility fail

I decided to check out some profiles. I right clicked someone’s username and clicked ‘Open in new tab’. I switched to the new tab. Blank page (with a URL of about:blank). What? I looked back on the original page. Look at the status bar when I hovered over a link:

See that “javascript:;” bit? That’s called inaccessibility. It means that Menshn makes use of javascript to show things like the popup box of a user’s profile. What it doesn’t do, though, is use the widely-approved progressive enhancement technique to make this javascript-powered link do something useful when the user isn’t using javascript (or indeed, just trying to open things in new windows). Twitter’s interface is hugely javascript-enhanced, but they don’t make this error. It wasn’t just here either — the entire site uses javascript like this, and all of the functions are defined in a big <script> block at the top of every page, polluting the global namespace. This might be nerd stuff, but it also means the site won’t scale well (try introducing advertising into that context) or work for users on limited devices.

We laugh at bandwidth

More nerd stuff: I ran Google’s PageSpeed tool against the site. It scored a paltry 56 / 100. High on the list of reasons for the poor score was the way it shows user avatars: the site doesn’t seem to resize the user’s uploaded avatar, so if the user uploads a massive image for their profile, Menshn simply downloads it and then displays it as a tiny size, making the user pay the download speed price. To illustrate, here’s the actual size of one user’s avatar (400×400 pixels) compared to the thumbnail size (48×48 pixels) Menshn shows it at — 98% smaller:

All the images suffer from this, with Google estimating it adding over half a megabyte to the page download weight. Poor.

Ego and arbitrariness

Menshn allows users to post updates of 180 characters — why? So it sounds better than Twitter with its paltry 140? Twitter’s limits were defined by SMS limits, as the service started out. Making it 180 is arbitrary and pointless. If you want to give users more text, give them significantly more, otherwise stick to what they’re already conditioned to work with. Although their //username notation adds an extra character over @username…

Finally, the homepage greets you with this at the bottom:

Every time I’ve checked the site, Bozier and Louise occupied the top spots. I assume this is because the site ranks them on their cumulative karma score (or whatever Menshn refers to this score as). What’s the point in a system like this? Showing the all-time highest rated users will end up with a boring, self-fulfilling list, as “power users” are continually highly rated purely because they already have a huge score. A smarter options would be to show users who accrued the most upvotes in a given time period, similarly to Randall Munroe’s elegant solution for reddit’s commenting system.

What I would have done differently

Well, the first point is that I wouldn’t have even started it — so perhaps that’s a plus point against Menshn for even bothering. Personally, though, aiming to compete against the giants is almost always pointless. Menshn might argue that they’re not competing or usurping Twitter et al, and they’re not (they couldn’t). What they have done, though, is built their own platform rather than taking advantage of the existing one. Twitter definitely does have issues with relevancy and filtering, and their Lists tool doesn’t really solve this. If Menshn was an app which used your existing Twitter account and allowed you to filter things into topics, I might have been genuinely interested. Inventing a brand new system with none of the appeal of Twitter and many added flaws, I see no real reason to invest my time in it.

Poor UX/design and some lack of frontend optimisation aren’t impassible problems, but they demonstrate a lack of awareness on how to build a compelling and scalable product. Menshn is possibly the seed of a good idea, but as of today, it’s not germinated enough for it to be useful.